Protein engineering usually involves the modification of genetically coded proteins by site-directed mutagenesis. As an alternative to this strategy, we are exploring the de novo design and chemical synthesis of nongenetic beta proteins. Our ultimate goal is to design and build nongenetic proteins containing a binding site (a receptor) or a catalytic site (an enzyme). Initial efforts will focus on two identical peptide chains held together by one or two covalent bonds, each designed to fold into an antiparallel beta-pleated sheet consisting of four beta strands connected by three beta turns. The first and third beta turns would constitute a potential active site at the open end of the barrel. Each beta sheet would have a hydrophilic face and a hydrophobic face. In aqueous solvent, this pair of amphiphilic sheets might adopt a tertiary structure having the two hydrophobic sides held together by noncovalent forces to give a small globular beta-barrel bell-shaped protein, which we call a betabellin. Alternatively, if a parallel or antiparallel beta sheet is constructed from beta strands attached to a symmetric cross linker, it might produce a small globular beta-cylindrical protein, which we call a betacylindrin. An early aim of this project is to engineer nongenetic proteins that fold into small globular structures, which may form crystals suitable for X-ray crystallography, such as crystalline betabellins having beta-barrel architecture or crystalline betacylindrins having beta-cylindrical architecture. A new and general procedure we propose for engineering of nongenetic proteins uses several applications of a four-step engineering cycle, namely, design of potentially useful variants by computer-aided molecular modeling, construction of many structural variants by combinatorial protein synthesis, selection of the desired functional variants by competition affinity chromatography, and identification of the selectyed variants by Edman sequencing. This procedure should allow the engineering of nongenetic receptor proteins by using appropriate ligands for the desired binding site in the selection step, or of nongenetic enzymes by using appropriate transition-state inhibitors of the desired catalytic site in the selection step. Intermediate goals include the engineering of betabellins with binding or catalytic sites, such as a triose phosphate isomerase, and engineering of betacylindrins with binding or catalytic sites, such as a fatty acid carrier. These tailor-made nongentic proteins may be useful for medical diagnosis or therapy.